Stochastic H∞ identification

an iteratively weighted least squares algorithm

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a novel problem formulation and algorithm for H system identification based on a stochastic noise model and constrained model set to reduce the conservatism in deterministic noise models, and statistical inefficiency and computational complexity associated with high-order estimates. By establishing a connection between a minimax problem and a sequence of weighted least square problems, we show that the proposed stochastic, constrained problem can be solved with a computationally attractive and conceptually simple iteratively weighted least square (IWLS) identification algorithm. The IWLS procedure is based on a sequence of standard parametric weighted least square output error identification routines, where the weighting is updated via non-parametric estimation of the modeling error to asymptotically achieve the H identification criterion.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE Conference on Decision and Control
PublisherIEEE
Pages3374-3379
Number of pages6
Volume4
StatePublished - 1994
EventProceedings of the 33rd IEEE Conference on Decision and Control. Part 1 (of 4) - Lake Buena Vista, FL, USA
Duration: Dec 14 1994Dec 16 1994

Other

OtherProceedings of the 33rd IEEE Conference on Decision and Control. Part 1 (of 4)
CityLake Buena Vista, FL, USA
Period12/14/9412/16/94

Fingerprint

Computational complexity
Identification (control systems)

ASJC Scopus subject areas

  • Chemical Health and Safety
  • Control and Systems Engineering
  • Safety, Risk, Reliability and Quality

Cite this

Rangan, S., & Ren, W. (1994). Stochastic H∞ identification: an iteratively weighted least squares algorithm. In Proceedings of the IEEE Conference on Decision and Control (Vol. 4, pp. 3374-3379). IEEE.

Stochastic H∞ identification : an iteratively weighted least squares algorithm. / Rangan, Sundeep; Ren, Wei.

Proceedings of the IEEE Conference on Decision and Control. Vol. 4 IEEE, 1994. p. 3374-3379.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rangan, S & Ren, W 1994, Stochastic H∞ identification: an iteratively weighted least squares algorithm. in Proceedings of the IEEE Conference on Decision and Control. vol. 4, IEEE, pp. 3374-3379, Proceedings of the 33rd IEEE Conference on Decision and Control. Part 1 (of 4), Lake Buena Vista, FL, USA, 12/14/94.
Rangan S, Ren W. Stochastic H∞ identification: an iteratively weighted least squares algorithm. In Proceedings of the IEEE Conference on Decision and Control. Vol. 4. IEEE. 1994. p. 3374-3379
Rangan, Sundeep ; Ren, Wei. / Stochastic H∞ identification : an iteratively weighted least squares algorithm. Proceedings of the IEEE Conference on Decision and Control. Vol. 4 IEEE, 1994. pp. 3374-3379
@inproceedings{97365b26ba93422dba1a7feef16dd4b2,
title = "Stochastic H∞ identification: an iteratively weighted least squares algorithm",
abstract = "We present a novel problem formulation and algorithm for H∞ system identification based on a stochastic noise model and constrained model set to reduce the conservatism in deterministic noise models, and statistical inefficiency and computational complexity associated with high-order estimates. By establishing a connection between a minimax problem and a sequence of weighted least square problems, we show that the proposed stochastic, constrained problem can be solved with a computationally attractive and conceptually simple iteratively weighted least square (IWLS) identification algorithm. The IWLS procedure is based on a sequence of standard parametric weighted least square output error identification routines, where the weighting is updated via non-parametric estimation of the modeling error to asymptotically achieve the H∞ identification criterion.",
author = "Sundeep Rangan and Wei Ren",
year = "1994",
language = "English (US)",
volume = "4",
pages = "3374--3379",
booktitle = "Proceedings of the IEEE Conference on Decision and Control",
publisher = "IEEE",

}

TY - GEN

T1 - Stochastic H∞ identification

T2 - an iteratively weighted least squares algorithm

AU - Rangan, Sundeep

AU - Ren, Wei

PY - 1994

Y1 - 1994

N2 - We present a novel problem formulation and algorithm for H∞ system identification based on a stochastic noise model and constrained model set to reduce the conservatism in deterministic noise models, and statistical inefficiency and computational complexity associated with high-order estimates. By establishing a connection between a minimax problem and a sequence of weighted least square problems, we show that the proposed stochastic, constrained problem can be solved with a computationally attractive and conceptually simple iteratively weighted least square (IWLS) identification algorithm. The IWLS procedure is based on a sequence of standard parametric weighted least square output error identification routines, where the weighting is updated via non-parametric estimation of the modeling error to asymptotically achieve the H∞ identification criterion.

AB - We present a novel problem formulation and algorithm for H∞ system identification based on a stochastic noise model and constrained model set to reduce the conservatism in deterministic noise models, and statistical inefficiency and computational complexity associated with high-order estimates. By establishing a connection between a minimax problem and a sequence of weighted least square problems, we show that the proposed stochastic, constrained problem can be solved with a computationally attractive and conceptually simple iteratively weighted least square (IWLS) identification algorithm. The IWLS procedure is based on a sequence of standard parametric weighted least square output error identification routines, where the weighting is updated via non-parametric estimation of the modeling error to asymptotically achieve the H∞ identification criterion.

UR - http://www.scopus.com/inward/record.url?scp=0028738674&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028738674&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

SP - 3374

EP - 3379

BT - Proceedings of the IEEE Conference on Decision and Control

PB - IEEE

ER -